Discovering New Sentiments from the Social Web
This addresses the challenge of understanding affective phenomena in digital social systems, but it appears incremental as it builds on existing sentiment analysis without presenting concrete results.
The paper tackles the problem of identifying novel sentiment structures that emerge in social networks, proposing that these hypercomplex sentiments cannot be captured by traditional human-defined categorizations.
A persistent challenge in Complex Systems (CS) research is the phenomenological reconstruction of systems from raw data. In order to face the problem, the use of sound features to reason on the system from data processing is a key step. In the specific case of complex societal systems, sentiment analysis allows to mirror (part of) the affective dimension. However it is not reasonable to think that individual sentiment categorization can encompass the new affective phenomena in digital social networks. The present papers addresses the problem of isolating sentiment concepts which emerge in social networks. In an analogy to Artificial Intelligent Singularity, we propose the study and analysis of these new complex sentiment structures and how they are similar to or diverge from classic conceptual structures associated to sentiment lexicons. The conjecture is that it is highly probable that hypercomplex sentiment structures -not explained with human categorizations- emerge from high dynamic social information networks. Roughly speaking, new sentiment can emerge from the new global nervous systems as it occurs in humans.